Authenticating Marriage: The Decree Tametsi in a Comparative Global Perspective^*

Global history takes two forms. The first is structured chronologically, the second comparatively. The focus of the (very broadly) chronological sort is on seminal developments and influences. Questions about how the West became more powerful and richer than the rest tend to dominate, but the approach also works for religious history, not least for the spread of Christianity as a world religion throughout the globe. Under this large rubric the history of the Council of Trent’s influence deserves a prominent position, for missionary work was almost a Catholic monopoly in the early modern period, and the effect of the Council of Trent’s decrees on the areas missionised, in general and with special reference to marriage, ought to be a major theme in that story. In fact, the proportion of historical importance to research undertaken is higher for this topic than for most, and most of the contributions to this Focus section help to redress that.

It is also possible, however, to study the Council of Trent’s marriage decrees in the framework of the other sort of global history, that is to say, comparatively. By looking at responses of different societies to approximately the same problem, one can begin to discern patterns of similarity but also of specificity. It is the aim of the present essay to ask such questions.

The Council of Trent attempted to change a situation in which couples could get married validly without any formalities. Here we look at that attempt in the light of other societies that had to face the same problem: medieval Europe, which addressed it wholly ineffectually; modern Pakistan, which has still to overcome a conviction aired in the discussions before the Council of Trent’s decree Tametsi on marriage: viz., that positive law lacks the power to impose an extra condition for the validity of marriage; and eighteenth-century England, which tried to solve the problem along lines somewhat similar to those laid down by the Council of Trent. It emerges forcibly that this was no easy problem to solve: that is common to all four cases. There is also a finding about the specificity of the Counter-Reformation solution. Difficult though the problems created unintentionally by Tametsi were, the creation of the Congregatio Concilii as the body with the task of implementing the Council’s non-dogmatic decrees provided a way of dealing with them. Administrative in character rather than legal in the ius commune sense, the Congregatio Concilii seems to have been relatively nimble in responding to and resolving the problems that new legislation invariably generates.

Under the aegis of Pope Innocent III, with the medieval papacy arguably at the peak of its power, in the context of across-the-board reforms, as then conceived, the Fourth Lateran Council of 1215¹ revolutionised marriage law in that it reduced the »forbidden degrees« within which a valid marriage could not be contracted from seven (sixth cousin) to four (third cousin) and required that banns be read before marriages, in the hope that impediments would be revealed before the knot was tied.² What Lateran IV did not do was make the prior reading of the banns, or any form of public ceremony or registration, a condition for a marriage to be actually valid – banns were required but breaking that rule would not be a ground for annulment. It left the rule established in preceding |decades that free consent was sufficient to make a marriage valid,³ and the consequences unfolded over the next centuries. Until the Counter-Reformation a couple could validly get married entirely on their own – without priest, magistrate, or even witnesses – in Catholic Europe: Church authorities did not like it, but they recognised the faits accomplis. Granted that Lateran IV changed so much, it is good to think about the major negative fact that the thirteenth-century council missed what might seem to the modern observer an excellent opportunity to anticipate the Council of Trent’s ruling on marriage in church: for it is well known that in the sixteenth century the Council of Trent made a priest’s presence a sine qua non for a valid marriage between Catholics (decree Tametsi).⁴ The later medieval marriage law’s unwilling acceptance of unauthenticated marriages is the point of contrast with three comparanda: modern Pakistan, eighteenth-century England, and Tridentine Catholicism, all of which faced the same problem and tried to fix it by making formal authentication procedures a sine qua non for a valid marriage.

The first part of this essay explores the consequences of the medieval Church’s inaction, but we then go on to ask what the alternatives were. Only comparative analyses can make this relatively concrete. They all show how hard it was to move from the assumption that divine law alone determined what constituted a marriage to the regulation of validity by positive human law – without breaking the link with the religious system, as modern Western states do. Comparison and contrast with modern Pakistan suggests that the Western Church was better placed to legitimate such a change at least with the religious elite, but also that religious assumptions about the minimum requirements for a valid marriage would not have been easy to alter, granted a widespread assumption that unofficial contracts were valid in the eyes of God. Eighteenth-century England suggests that even an apparently well-crafted law could create entirely unforeseen problems: thus, even if the Fourth Lateran Council had passed such a law, they would have faced a host of unforeseen problems. The aftermath of the Council of Trent also makes the same point, which is brought home by soundings in a little-known fondo of the Congregation set up to implement Tridentine decrees. The same soundings also show a method of resolving them quite different from that of classical medieval canon law, and raise the question of whether this wider range of changes was even conceivable in the thirteenth century.

The comparative method used here is combined with a certain sort of counterfactual analysis. It is a different sort from speculations about what would have happened if Napoleon had won the battle of Waterloo, and equally far from the attempts at counterfactual quantification of the »New Economic History« that was in vogue especially in the 1960s and 1970s. It is closer to the approach of Geoffrey Hawthorn,⁵ whose method deserves more sustained attention from historians. Hawthorn’s method was to ask whether a given counterfactual scenario was within the realm of plausibility at a given time. Did people making decisions at the time have enough room for manoeuvre to bring about this other outcome? He argues, for instance, that the Black Death was unstoppable by human agency when it first hit Europe in 1348. A couple of generations later, however, people knew that a really ruthless cordon sanitaire could keep it away from a given city or region. There was an alternative, though it required harsh decisions. Where a regime was in place capable of enforcing such decisions, but did not do so, a different outcome, escape from the plague, is plausibly counterfactual. The objection to some counterfactual statements, that they do not answer historical questions, need not hold with Hawthorn’s methodology or with the variant of it employed here. We offer modal statements about the past in answer to modal |historical questions,⁶ defining modal as »involving the affirmation or negation of possibility, impossibility, necessity, or contingency« (to quote the Oxford English Dictionary).

The variant of Hawthorn’s approach adopted here is to explore the range of other plausible outcomes with the help of the comparative method, by investigating the ways in which three societies tried to confront and surmount the problem that the Western Church failed to face in 1215, at the Fourth Lateran Council. The concrete comparisons make it easier to envisage the other paths that were not taken in 1215. An essential basis for the comparison is a common starting point in all four cases: comparison has to be carefully framed to match like with like. In all the cases studied, marriage is regarded as a religious institution, and the legislative body – a great medieval Church Council, an Islamic State, the eighteenth-century Crown in Parliament, and an early modern Church Council – treats this institution as within its purview.

The comparisons show us that the alternative to tolerating one set of problems was to generate a different set, which we can characterise with some specificity. The comparanda adopted here do not exhaust all the possible ways of attempting to deal with the problem facing the medieval church, but they do bring out in their different ways how elusive the solution to the problem was, unless – and it is the final comparison that brings this out – change in marriage law had been accompanied by fundamental changes in the Church’s decision-making system.

In the centuries between the Fourth Lateran Council in 1215 and the Council of Trent (1545– 1563) the marriage system of the medieval Church achieved in some respects a high degree of rationality, if we define the term to mean internal consistency governed by general principles. The internal contradiction by which the rules about consanguinity and affinity were used to undermine the principle of indissolubility had been more or less resolved.⁷ A positive law of marriage – that is, a man-made law designed to implement divine principles, but not coterminous with them – was included in the easily accessible corpus of canon law (the law of the medieval Church), and a network of courts all over Europe was staffed by men who knew enough to apply it.⁸ Papal justice could reach all over Europe through a system of judges delegate,⁹ and could be meticulous about the formal rules of procedural law.¹⁰

On the particular point that concerns us, however, the social outcomes of the system were highly irrational, in the sense that they were chaotic and contrary to the intentions of those who had made and who administered the law. It was often hard to be sure whether or not a couple had been married, if the alleged consent had been exchanged in informal circumstances and if one partner had subsequently married someone else. The Church deeply disapproved of clandestine marriages and in some regions at least the spiritual penalties for them were severe, but these contracts were recognised as true marriages rather as a mass said by a priest in mortal sin was recognised as valid.

Marriage customs varied widely across Europe, and in parts of Italy, for instance, it was quite normal to get married without a priest or religious ceremony: this did not necessarily imply any disrespect for the Church, since the Fourth Lateran Council made no regulations at all about a church ceremony,¹¹ as opposed to banns before marriage. It should be said, however, that even the rule about the banns seems to have been ignored in some |regions, not only by disobedient lay couples, of which more below, but also apparently without the local ecclesiastical authorities making an issue of it.¹²

Quite apart from legitimate local custom where the actual ceremony was concerned, and even apart from the failure of some regional churches to enforce the Lateran IV rule about banns, the situation was almost out of control, due precisely to this lack of any system for officially authenticating marriages. Without the help of such a system, the ecclesiastical courts that had come to maturity in the twelfth and thirteenth centuries constantly had to decide whether or not a marriage was valid. The consequence was a mass of tricky litigation in the ecclesiastical courts. Cases about alleged prior contracts were extremely common in the church courts of the later Middle Ages and have been well studied by historians – notably by Richard Helmholz (the pathbreaking pioneer),¹³ Frederik Pedersen¹⁴ and by Charles Donahue in a recent monumental study of five courts.¹⁵ For the law was open to abuse. At the cost of some perjury, the law that all marriages, even clandestine ones, were unbreakable could be turned into a way of breaking a marriage, by the type of suit known as the »marriage and divorce case«, causa matrimonialis et divorcii.¹⁶ The husband or wife who wanted to get out of a marriage in order to marry a third party could invent a story of a clandestine marriage to the third party, one which allegedly took place before the marriage that he or she was attempting to get annulled. If witnesses could be induced by friendship or cash to swear to a plausible story, the court might be convinced and declare the existing marriage to be null. Obviously, this could only work if the age of the third party did not rule it out: a man who had been married for twenty years could not claim that he had married a twenty-year-old girl before that!

It must have been possible to abuse the system in a different way: by using it for purposes of seduction. Helmholz discusses a case that came before Innocent III, in which a man »contracted marriage under a fictitious name in order to extort sexual intercourse from a girl«.¹⁷ Innocent decided that the marriage was invalid if the man simply used the contract as a trick without any internal intention of making it a marriage. On the other hand, a consensus developed that church courts should uphold such marriages:¹⁸ the ecclesiastical court could only annul a marriage on the basis of proof. A man’s declaration that he had not meant what he had said could not be grounds for an annulment. On the other hand, a marriage’s existence or non-existence was an objective fact which the court’s judgement could not affect.

The situation could easily arise where the church court upheld a marriage which one party knew to be invalid. If the person in question had married someone else with real consent after the marriage based on feigned consent the situation was worse. The other partner to the marriage which had been upheld by the court could demand re|stitution of conjugal rights, but to comply would be to commit adultery.¹⁹ Or again: Helmholz mentions the case of a man called John Paynaminuta living in London whose first wife was seen alive in Bayeux. Even if he wanted to leave the second wife, to do so legally he had to somehow prove to a church court that it really was his first wife in Bayeux, which might be hard.²⁰

The further rule that betrothal followed by sexual intercourse counted as evidence for consent to a marriage must also have been a potential source of ambiguity; and it could have generated situations similar to those outlined above. The intercourse did not actually constitute consent: it was merely thought normally to be overwhelming evidence of consent when preceded by an engagement. As Aquinas puts it:

[…] we can speak in two ways about marriage. In one way with respect to the forum of conscience; and in this way the truth of the matter is that carnal union does not have the power to perfect a marriage which has been preceded by an engagement in words of the future tense, if mental consent is lacking; since even words in the present tense expressing consent do not make a marriage if mental consent is lacking […]. In another way with respect to the judgement of the Church: and since in an external judgement, judgement is passed according to what can be seen from the outside, carnal union following betrothal is judged to make a marriage according to the judgement of the Church unless some explicit signs of trickery or fraud are apparent.²¹

A little later he specifies »signs […] such as if they are far apart in status, either with respect to nobility, or to fortune«. These comments evoke Tess of the d’Urbervilles-like scenarios – seduction of simple girls by cynical upper-class men. There was not a lot to be done about them unless the law was modified.

The law did not work in a uniform way all over Europe: in fact, there was much variety. In the bishopric of Regensburg, for instance, there was a much more tolerant attitude to informal marriages than, say, in Salisbury.²² The magisterial study of marriage litigation by Charles Donahue brings out some striking regional differences between England, Paris, and Belgium, the areas he studied.²³ His arguments are too substantial for précis here but the following schema may be risked: the more that family interests controlled marriage at point of entry, the smaller the room left for exploiting the openings left by canon law. The converse is to be expected where the extended family was relatively weak and social attitudes relatively »individualistic«: the easier it was to get married without prior public knowledge, the higher the risk that one partner might deny that the marriage had ever taken place and get away with it, and the greater the temptation to escape from a marriage by pleading a previous contract and demonstrating it to a court through friends prepared to perjure themselves. On this model, parental or extended family control, on the one hand, and individualistic anarchy, on the other, varied inversely.

For all the regional variety, the absence of an official authentication system was a problem common to the whole Latin West. Family structures might vary; local church authorities might be laissez-faire and lenient, or tough and prescriptive; the fact remains that a couple who had reached puberty and who were not too closely related could get married at any time and place by exchanging words of free consent. This was not likely to please parents any more than it did the authorities, and in due course there was a powerful reaction against it.

In the sixteenth century, forces as different as the French monarchy and Martin Luther wanted parental control to be enforced by law. If Sarah Hanley is right, mid-sixteenth-century French juries were asserting the right not only to punish marriages contracted without parental consent but to actually annul them, as well – a very different |matter.²⁴ A French royal edict of 1556 »forbade young men under the age of thirty and young women under twenty-five to marry without obtaining the consent of their parents or relatives«, with severe penalties for disobedience: loss of inheritance and other property rights (e.g., gifts might be taken back),²⁵ and »peines arbitraries«.²⁶ The Edict of Blois laid down that to marry someone under twenty-five against their family’s wishes was rape and punishable by death; the consent of the »minor« did not alter the case.²⁷

To make parental consent a necessary condition for a valid marriage was one answer to the problem. To go along with this idea the Church would have had to reverse a long tradition of emphasis on individual choice, especially pronounced from the twelfth century onwards.²⁸ At Trent, the Council Fathers were not prepared to do so, despite strong pressure from the French king to make parental consent a condition for a marriage of minors (»filii et filiae familias«).²⁹

Luther seems to have thought that lack of parental consent invalidated a marriage, and came to the view that even subsequent parental consent did not make it a real marriage.³⁰ A partial concession to the medieval canonists’ point of view was his belief that subsequent consummation created a valid marriage.³¹ His position is complicated by his use of the word »Verlöbnis«, which one would normally translate as »betrothal«. Does he mean a promise to marry in the future, or present consent? The solution appears to be that he collapsed the distinction between betrothal and marriage so that all betrothals were marriages unless the contract was explicitly subject to a condition or a time limit.³² Luther’s marriage doctrine does not lend itself to simple summary.³³ His views are, however, certainly symptomatic of a widespread sense that marriage by consent alone was not working.

The actual solutions worked out in Protestant Germany must be left to one side here. It is hard to generalise about a country with many different |regimes, and more research is needed to put together an overall interpretation.³⁴ In any case three comparanda are enough. In each of the cases chosen, the problem facing the authorities was practically the same as with the medieval Church, but they chose action rather than inaction and in different ways show the range and limits of what the medieval Church could have done about the problem.

All three systems to be compared with that of the late medieval church attempted a rationalisation³⁵ of entry into marriage through the imposition of a clear-cut system under public control. The fact that the public control is that of the State in two cases and the Catholic Church in the other two makes surprisingly little difference: neither modern Pakistan nor eighteenth-century England recognised a Church – State or even a Religious – Secular distinction such as non-academic Westerners believe to be universal, and the medieval and early modern Church acted as »the government« where the validity of marriage was concerned, getting its claims fairly generally accepted in Catholic Europe, moreover, with the notable exception, to be discussed below, of France.

So, in each of these three cases – modern Pakistan, eighteenth-century England, Counter-Reformation Catholicism – a body claiming religious authority within the system in question tried to put an end to the chaos arising from religious recognition of informal marriages by requiring official authentication as a condition of validity. In each case the authority worked within the religious system rather than against it; so we are not looking at anything like a Church v. State or Religious – Secular conflict. In each case the solution brought its own problems, though in one case a mechanism was devised for solving them relatively smoothly. We will begin at a distance and in the present, with modern Pakistan; come closer to medieval Europe chronologically and geographically by looking at the abolition of clandestine marriages in eighteenth-century England; and end back in sixteenth-century Catholic Church, with the Council of Trent’s solution to the problem of unauthorised marriages.

Modern Pakistan is an Islamic state³⁶ which has found itself facing difficulties of much the same sort as troubled the later medieval Church: it was and is difficult to tell who is married and who is not. Marriages unauthenticated by any official body were of unquestioned validity in religious terms – provided one could tell who was married to whom. To introduce some order, a body claiming supreme authority within a system of sacred law, as well as over State law, which is officially integrated into the sacred law, has purported to declare that unregistered marriages are invalid. Against this, many people doubt whether a marriage can be invalidated if it fulfils the essential requirements of the Sharia’h law. Pakistan has taken the course that the Fourth Lateran Council implicitly rejected but which, as we shall see, the Council of Trent adopted. The fact that we are dealing with Islamic rather than Christian sacred law does not invalidate the comparison: it works at all the essential points.

Can a Muslim state make Muslim laws and put itself above God’s law? Pakistan is the one jurisdiction in the world which has had to determine such questions within a postcolonial legal framework of reference. It becomes a matter of defining the identity of the Islamic Republic of Pakistan how questions of marriage, divorce and other matters are decided by state courts. When Pakistan became independent in 1947 it aimed to provide a homeland for the subcontinent’s Muslims. A national Convention decided already in 1949 that the basic principles of Pakistani law should be those of Islam. This set up a tricky predicament for Pakistani state courts: the judges have to do the job of traditional Muslim jurists.|

In 1961 the Muslim Family Laws Ordinance (MFLO) was promulgated.³⁷ This is what section 5 says about the contract of marriage:

5. Registration of marriages:(1) Every marriage solemnised under Muslim Law shall be registered in accordance with the provisions of this Ordinance.

What does this mean in practice and in law? The MFLO evidently recognises that a Muslim marriage will be celebrated first of all under Muslim law, as a solemn contract before Allah. The Muslim state of Pakistan could not possibly have disregarded this. But what is the relationship between the traditional Muslim family law, the shariat, and the state-made rule of law that says that such a marriage shall be registered under the legal provisions made by the state? When does a Muslim marriage in Pakistan become legally valid – on completion of the Muslim contract, or on registration of the marriage in accordance with state law? This means we are asking whether the state’s law or God’s law is superior.

If we read the provision again, it seems to give us an answer: section 5 (1) avoids the word »contracted« for marriage, but uses »solemnised«, which seems to suggest that a religious marriage will only be legally valid under Pakistani law if it is duly registered with the state authorities. This impression is also reinforced by the use of the word »shall« for the requirement to register.

Does it mean that a Muslim marriage contract in Pakistan will only be legally valid if that contract is ratified by the state? By no means! There is not a single case from Pakistan that would tell you that an unregistered Muslim marriage is not legally valid.³⁸ The MFLO has always been challenged by Muslim fundamentalists in Pakistan as an un-Islamic law, and in 2000, the Federal Shariat Court of Pakistan finally tried to resolve the many challenges to the MFLO. It undertook a national reconciliation exercise, as the Court went to all major cities in the country, inviting anyone with any views to come and tell them, so that the matter could be decided in the public interest, helping to construct a proper Islamic system of governance in Pakistan.

The case of Allah Rakha v. Federation of Pakistan, PLD 2000 FSC 1, therefore became candid proof that Pakistan has two types of Islamic law and that, if it comes to the crunch, Islamic natural law is superior to Islamic positivist law.³⁹ The case concerns succession of orphaned grandchildren, marriage, bigamy, and divorce. We are only looking here at the marriage aspects, sufficient to give the essential picture. The Federal Shariat Court was told by certain petitioners that the requirement in s. 5 MFLO to register Muslim marriages under Pakistani law was un-Islamic and violated the Qur’an and Sunna. With considerable subtlety, the Court managed to hide the key question at issue here: Is a traditional Muslim marriage contract entered into in Pakistan legally valid by itself, or does it need the sanction of the state? It is a clear question: cannot the Court give a clear answer? Section 5 (1) seems to suggest that a marriage must be registered, but we have already seen that if we read »shall« as »must«, we are privileging state law over shariat law, and that would indeed be un-Islamic. The Federal Shariat Court gets out of this: they find that there is nothing in Qur’an and Sunna to say that a marriage should not be registered. So, a legal requirement to register a marriage would not in itself be un-Islamic, and in fact the Sunna recommends that important transactions should be documented in writing.

That line of reasoning does not, however, solve the issue of when a Muslim marriage contract becomes legally binding. In fact, where there is a conflict between the state-sanctioned shariat law and the state-made MFLO, the rules of the shariat will always prevail. Thus, »shall« in the MFLO does not mean »must«, but rather something like »should«, or »should please«, or maybe »Inshallah, if God wills«. The result is that the so-called legal requirement to register a Muslim marriage in Pakistan under the provisions of the MFLO is in fact optional. As in the early days of Islam, a modern Islamic state is also today not in a position to override God’s law. If the simple rule of contract of marriage in Islamic law is that a man |and a woman may contract a marriage in front of God and at least two witnesses, then no Islamic state can deny that contract legal validity on the ground that it does not comply with some formalism of state law. Thus, in Pakistan, we have a soi-disant Islamic government laying down an extra condition for a valid marriage, but failing to carry weight with the learned men who mattered and the general population.

Is this what would have happened if the Fourth Lateran Council had tried to make some kind of registration of marriage a requirement for validity? The question forces us to think about authority structures in the two religions but also about the relation between those structures and popular feeling. On the one hand, it seems likely that the requirement could have been legitimated in the eyes of the religious elite in thirteenth-century Western Europe. On the other hand, it is not so clear that the elite could have carried popular feeling with them.

The religious elite would probably have accepted that a General Council had the power to impose a new condition for validity, because there were precedents. The Second Lateran Council (1139) had declared that a priest could not get married validly.⁴⁰ Furthermore, the changing of the boundaries of the forbidden degrees at Lateran IV implied that impediments to a valid marriage existing up until that point had been positive, i.e., human law, which the Church had made and could unmake. Subsequent dispensation practice presupposed that the impediments removed were those that the Church itself had created – and these were impediments that invalidated a marriage.⁴¹ Logically and theologically, no insuperable theological obstacle stood in the way, if the will had been there. If the Fourth Lateran Council had decreed (say) that no marriage was valid unless registered by the parish priest or a notary, it seems likely that the Council’s own authority would have given the decision more authority than the Pakistani State’s law seems to have had. However Islamic it may claim to be, the Pakistani State does not possess within Islam (even within the borders of Pakistan) the kind of authority that a General Council possesses within Catholic Christianity.

Whether the population of Christendom would have understood the logic is another matter. We have seen that there were regions of Europe where the clergy were not even involved in the marriage ceremony – where such involvement had not even been required by the Church so far as the actual ceremony was concerned. It is doubtful whether it was high time to win acceptance for so drastic a change.

When the change was finally made in the sixteenth century by the Council of Trent the atmosphere was different. By that time, the clerical elite could at least draw support from the patriarchal reaction to the »consent makes a marriage« anarchy of the later medieval centuries. While the Council Fathers parted company from the French monarchy and Martin Luther, among many others, when it came to parental consent, they could at least agree with both about the danger of clandestine marriages. In the thirteenth century that patriarchal backing might have been lacking. The power that the consent doctrine gave to couples had not yet had time to penetrate the general consciousness and subvert family authority. The legitimation by the highest Church authority of the doctrine that consent between a couple made a marriage, and the enforcement of that doctrine through a network of Church courts, was still relatively new.⁴² Potentially it put power in the hands of young people to defy their parents’ wishes. A general consciousness of that power would have taken time to take hold. It had well and truly done so by the sixteenth century, and as we have seen this had provoked a reaction. The reaction surely softened public opinion and made it ready for marriage reform by the Church. Heads of households in particular must by then have been acutely aware of the advantages of eliminating informal entry into marriage. Parental control would be best of all but proper procedures with a religious ceremony were a lot better than nothing: there would be every opportunity to counter romantic plans. In 1215 the Church could hardly have counted on the same kind of support. We cannot assume that the Fourth Lateran Council would have been able to do what the Council of Trent just about managed to achieve. |

Comparison with Islam draws attention to the problem of public opinion. Our next comparison considers the difficulty of drafting legislation to reform a marriage law, which was descended from, and more or less the same as, that of the medieval Church. This case, in eighteenth-century England, shows that even if public opinion was not a real problem, it was still extremely difficult to frame a law of authentication in such a way as to avoid paradoxical results.

In England⁴³ the marriage law of the Catholic Middle Ages survived the Henrician and Edwardian Reformations, so that we find it still operating, entirely recognisable, in the first half of the eighteenth century. It was only finally abolished in 1753 by the Hardwicke marriage act.⁴⁴ This made marriage by a clergyman necessary for validity,⁴⁵ with certain other strict conditions attached to give parents and guardians their due. In its intentions, the act seems remarkably in tune with the religious and moral values of its time. Reading the 1753 Act without the benefit of hindsight, one would think it to be watertight. Clause XI, notably, looks like a careful piece of drafting:

And it is hereby further enacted, That all Marriages solemnised by Licence, […] where either of the Parties, not being a Widower or Widow, shall be under the Age of Twenty one Years, which shall be had without the Consent of the Father of such of the Parties, so under Age (if then living) first had and obtained, or if dead, of the Guardian or Guardians of the Person of the Party so under Age, lawfully appointed, or One of them; and in case there shall be no such Guardian or Guardians, then of the Mother (if living and unmarried) or if there shall be no Mother living and unmarried, then of a Guardian or Guardians of the Person appointed by the Court of Chancery, shall be absolutely null and void to all Intents and Purposes whatsoever.⁴⁶

Who could have foretold the trouble to which this clause would lead? The passionate attempt to reform the Act by Dr Joseph Phillimore nearly three-quarters of a century later⁴⁷ shows how even most carefully framed acts could generate paradoxical legal situations. Just like the previous marriage law, it was open to abuse, which the courts had little choice but to endorse. For it appears to have been easy to give a false age when obtaining a licence. If someone pretended to be over 21 when they were not, there was nothing to stop them marrying without parental consent. If they subsequently revealed the truth about his age, the marriage was automatically void, and they were free to marry again.⁴⁸ Phillimore argues that men and women commonly used this rule to get out of a marriage long after it had taken place.⁴⁹ He gives examples, for instance Wattle v. Hathaway, where:

the woman was a minor at the time of the marriage; the husband obtained the license by making oath that she was of age. They cohabited some years, and had issue four children; when being in great poverty and distress he went to India, and there realised a considerable fortune. He returned to England [col. 1338]; and, after his marriage had subsisted twenty-seven years, instituted a suit for a nullity of the marriage, on the ground of his wife having been a minor at the time he had sworn her to be of age, and he succeeded in his suit.⁵⁰

Another kind of problem arose. What was the status of the consent of an illegitimate child’s |father? It should be remembered that in English law marriage did not retroactively legitimise a child: if a child was born out of wedlock, it was a bastard even if the couple were joined in matrimony shortly afterwards. Courts decided by legal logic that an illegitimate person’s father’s consent did not meet the requirements of the Act. Instead, a guardian had to be appointed by the High Court of Chancery.⁵¹ Only such a guardian could give the requisite consent. Logical this may have been, but it could produce a bizarre chain reaction: invalidating one marriage after another within the same family without anyone intending it. He gives as an example of a real case where two minors married with full parental consent, without knowing that the man had been born before his parents married. The couple had children, one of whom married as a minor with her father’s consent. When his own illegitimacy came out, his own marriage was invalidated because his natural father was not entitled to give consent. Consequently, his own consent to his daughter’s marriage was moreover without legal force, so it would be no marriage and her children too would be bastards.⁵²

Our second comparandum thus shows that unpredictable cases were likely to arise out of any attempt to create a rational law of marriage registration. The Hardwicke Marriage Act was passed two centuries after Trent’s attempt to remedy the abuses of clandestine marriage, and presumably with the benefit of that vicarious experience.⁵³ The point here is different from the Pakistani case. The authority of the Crown in Parliament over the Church of England was broadly accepted – unlike the authority of the Pakistani State over shariat law. Winning public acceptance of the law’s legitimacy was probably not a major problem. In eighteenth-century England, Church and State were united quite effectively under one sovereign body. Even so, Parliament’s attempt to put a working law in place regarding clandestine marriage still opened a can of worms. It would hardly have been different with the Fourth Lateran Council, even if it had been sufficiently sure of its power over the fundamentals of marriage to take any such step. It is true that the illegitimacy paradoxes were less likely to arise in Catholic canon law, in which bastards were legitimised by subsequent marriage of the parents. Even so, the Council of Trent’s reform of the medieval system also led to hard cases unforeseen by the Council Fathers.

Our third comparandum reinforces the argument of the second: it shows how hard it is to frame a law of authentication that really works, while satisfying religious requirements. The problems that arose out of the new law created by Trent have not yet been adequately studied.⁵⁴ A few scholars have started to trace the effort to make |the new decree work »on the ground«.⁵⁵ This effort has been handicapped by the unfamiliarity of the scholarly community as a whole, again with some exceptions, with a crucial source for studying hitches in the implementation of the new legislation.⁵⁶ This is the archive of the Congregation of the Council, the curial body headed by cardinals which was entrusted with the task of interpreting the reform legislation of the Council.⁵⁷ Some examples from this archive reinforce the point that it is hard to make registration necessary under a sacred law regime; this even with general acceptance of the Council’s power to make all future unregistered marriages null.

It would have taken an astute historian to predict from the decree Tametsi alone the problems to which it would give rise over the centuries. In a nineteenth-century compilation of the decisions of the Congregation, presented in a summary form without the documentation that lay behind them, they take up 110 pages.⁵⁸ These pages are full of human interest. Readers of Manzoni’s I promessi sposi may remember that when the parish priest decided to delay a wedding (after being threatened by a local noble who had designs on the bride), the impetuous bridegroom tried to trap him unawares into witnessing an exchange of vows.⁵⁹ The Congregation actually had to decide whether a valid marriage could emerge from such a scenario, and determined that the marriage should stand.⁶⁰

Another problem was: did the new rules hold good for France? Tametsi only applied to regions where the decisions of the Council of Trent had been promulgated. The French monarch prevented |its formal promulgation in France.⁶¹ There was a tendency in France for the monarchy and secular law to assert control over marriage.⁶²

The seriousness of the problem for contemporaries and its importance for historians of modern France can easily be underestimated. True, the French monarchy demanded marriage in Church, just like the Council of Trent; and it is also true that a priest who ignored the secular rule demanding parental consent was taking such a personal risk by defying the regime that marriages valid by Tridentine rules which were not also valid according to royal jurisprudence were unlikely to occur: that is to say, any marriage in front of the parish priest was likely to be a marriage for which parental consent had been obtained.

But a crucial point needs to be made here. If Tridentine rules did not apply in France, in the eyes of the Church, then marriages by consent alone would still be valid so far as canon law was concerned. The old medieval rules would still apply. The State could hardly prevent such marriages. It was one thing to put the frighteners on the parish clergy, and quite another to remove all possibility of personal and private exchanges of consent, below the radar of officialdom. In the eyes of the Church, however, such clandestine marriages would, if the Council of Trent was not canon law in France, invalidate any subsequent marriage in Church, even if it otherwise met all the requirements of Church and State. The medieval situation would still obtain, with all the power of clandestine marriages to out-trump any later ceremony. This consideration rules out the hypothesis with which some historians may have been implicitly content, viz. that the Council of Trent’s rules did not apply, but that this hardly mattered. For if it really were the case that they did not apply, so that clandestine marriages remained theologically valid, that would matter a great deal. Banns were read before marriages. If clandestine marriages were valid in ecclesiastical eyes, there would have been nothing to stop jilted spouses putting a spoke in the wheel of any subsequent, and more regular, marriage projects of their partner by telling the priest about the clandestine marriage.

Modern scholarship appears to have skirted around this major issue up until now. This may be because the Congregation of the Council has received too little attention. The decision seems to have been taken by the Congregation that Tametsi did apply there.

The key fact is that at the beginning of the seventeenth century the Congregation decided that promulgation of the Council of Trent was to be presumed in any parish where that decree (Tametsi) had been in fact observed as a decree of the Council of Trent – i.e., the parish’s practice was the litmus test.⁶³ From then on the decision was available in the Congregation’s archives for reference purposes. Centuries later, when France was in the throes of revolution, and many of the parish clergy were compromised by their adherence to the »Civil Constitution of the Clergy« (which had been condemned by the pope), we find this decision being cited in a letter from Rome to the bishop of Luçon, who had written asking what to do.⁶⁴ The bishop had proposed, as a solution to be considered, the line that since it could not be established with certainty that the |Council of Trent had been promulgated in France, its rules were not binding.⁶⁵ The »select committee« of cardinals given the task of replying to the bishop would have none of his ingenious suggestion and referred to, precisely, the Congregation’s archival record of its decision, and to its exact date, 26 September 1602.⁶⁶

There is much research to be done on the reasoning behind such decisions, which seem to have been the result of rational exchanges of opinion. An instance may be cited to give an idea of the kind of material to be found in the Congregation of the Council fondo. Tametsi had specified that the couple should be married by the parish priest. But suppose a family not only had a town house but also a pied-à-terre in the country? What if the marriage were conducted by the parish priest of their second dwelling? Such a case occurred at Florence. The case needs more investigation, but it would seem to have been initiated by the woman, who proposed that the marriage was null on the grounds that the real parish was the urban one.⁶⁷ Presumably she had changed her mind about the man. The kinds of consideration |that might affect the judgement are laid out: how often and for how long the family went to their country place; whether they kept the Florence house open when they did so, whether the country place was kept in good order continuously, where they did their Easter duties of confession and communion; the implications for many marriages of the precedent that would be set by annulling this marriage are noted; it is suggested that the tribunal of the Rota be consulted before the Congregation of the Council reaches a decision on how the case stands in relation to decrees of the Council of Trent.⁶⁸

A letter of enquiry was sent out to the Archbishop of Florence to ascertain the facts.⁶⁹ In the end, the Congregation’s decision seems to have been that marriages conducted by the parish priest of country cottages which were not a principal residence did not meet the requirements for a valid marriage as laid down by Tametsi.⁷⁰

All three of our comparanda show how difficult it is to devise a working law of marriage registration within a system where marriage also has religious significance. In a system of purely positive law, that is, where law is simply what the Sovereign State calls law and where the legislative sea is clear |of the rocks and currents of norms and values, rules for authenticating marriage become much easier. Even so, the current controversies about gay marriage, and the controversies which soon should be expected (in countries with large Islamic populations) about polygamous marriage, show that few modern legislatures can count on such easy sailing. The modern Pakistani State, the eighteenth-century English parliament, and the General Councils of the thirteenth and the sixteenth centuries all had to work around strong religious value systems, and we have seen they did not have it easy. The Fourth Lateran Council took the line of least resistance and did not tie authentication to validity: the system of banns was fine in principle but did not always work because marriages outside these rules were recognised as valid. The Pakistani law’s power to bind is contested and its religious status full of ambiguity. England continued the medieval system until the eighteenth century, then created a new law that produced some egregious paradoxes quite unintended by the legislators. The Council of Trent’s new law threw up a host of problematic situations equally unforeseen when the law was passed.

That said, this last comparandum stands out in that we find not only a host of problems but also a mechanism for dealing with them relatively efficaciously. The Pakistani State’s law seems quietly ineffective against the authority of a multitude of religious legists. The Hardwicke Marriage Act could not be reformed without a further act of Parliament, with all that this entailed. The Congregation of the Council, on the other hand, dealt with hard cases administratively so to speak, though after consultation with theologians or lawyers. The Congregation deserves a much more prominent place than anyone has given it in the history of Sacred Law. Its workings, as we have observed them above, fit very well with a thesis of Max Weber’s about papal history that gets less attention than his famous and more questionable argument about Protestantism and capitalism, viz. that:

everything was subject to the control of the central administrative bodies of the Curia, and the elaboration of binding ethical social norms could only take place via the highly flexible directives of these bodies. In this way a unique relationship between sacred and secular law grew up: namely that canon law actually became for secular law one of its guides on the path to rationality. And this was a result of the rational »institutional« character, of the Catholic Church, with which nothing comparable can be found in world history.⁷¹

Weber’s reference to the »Central Administrative Bodies« (die zentralen Behörden) of the Curia might be an allusion to the system of Congregations set up by the Counter-Reformation papacy.⁷² It is not likely that Weber knew much if anything about the workings of the Congregation of the Council but he may have been aware that it (and other »Congregations« set up during the sixteenth-century reform of the papal curia) made legal decisions.⁷³ At any rate, his general description fits the developments outlined in this essay uncannily well: not least their uniqueness, which a series of comparisons has brought into sharp relief. Parayre described the Congregation of the Council as an »Organe originale d’une administration sans modèle et sans copie«, adapting itself »avec aisance aux mille besoins d’une société cosmopolite«,⁷⁴ and the Positiones fondo of the Congregation, which scholars are only just beginning to explore, tends to bear out this striking claim. Of course, it should not be forgotten that at the level of local tribunals there were many inconsistencies between practice and the underlying principles of the system, as an eighteenth-century attempt at reform (by Benedict XIV) makes clear.⁷⁵ Nor were his efforts crowned with success, at least in the short term. That said, |it remains true that at the apex of the system, the Congregation took much of the chaos out of the administration of values, shared by the medieval and Counter-Reformation Church, which had been constantly frustrated by clandestine marriages which it had to recognise.

We have come to a point where the counterfactual comparative method with which we began has transcended the original question about what might have happened if clandestine marriages had been invalidated after 1215. The original question has indeed been answered. It is fairly clear that one way or another a host of problems would have arisen, as they did when such a step was taken in modern Pakistan, in eighteenth-century England, and at the Council of Trent: most of these problems could hardly have been foreseen in detail, even if the Council Fathers of the Lateran sensed them in outline. They could hardly have found a neat and clinical solution, though things would have been easier if something like the Congregation of the Council had been set up three and a half centuries earlier. The Congregation’s archive is testimony to both the problems generated by the Tridentine solution and a mechanism for fixing them.

The original question about the later Middle Ages has thus thrown the spotlight on a remarkable post-medieval institution. Almost in passing we have been able to provide new data on the Congregation of the Council’s decision on whether Tridentine marriage law should be deemed to have been promulgated in France, and we have been drawn into post-Tridentine intellectual discussions, preparatory to decisions, of a sort that historians would give anything to have as background to medieval decretals. Such shifts in focus should be welcomed in comparative counterfactual history, for it may often happen that a question that preoccupies historians of one period gets too little attention from historians of another, and that a comparative investigation originally intended primarily to illuminate one society ends up advancing our understanding of another society with which it has been compared. To make the most of this methodology historians have to risk venturing far beyond their personal comfort zones. The pay-off is greater dialogue between different sectors of the profession and, often, unexpected discoveries.